f

flag_videos_to_keep.py

@@ -0,0 +1,62 @@
+import json
+import shutil
+import os
+source_path = '/srv/ftp/hummingbird/2021'
+#target_path = '/home/thebears/Videos/ftp'
+target_path = '/home/thebears/ftp_links'
+
+
+
+have_json = set()
+for di, _, fns in os.walk(source_path):
+    for fn in fns:
+        if fn.endswith('.json'):
+            have_json.add(os.path.join(di, fn))
+
+
+
+
+
+
+do_stop = False
+fracs = dict()
+
+for c_js in have_json:
+    hits = 0
+    total = 0
+    o = json.load(open(c_js,'r'))
+    for i in o:
+        total += 1
+#        if len([x for x in i['scores'] if x > 0.1]) > 0:
+        if len(i['boxes']) > 0:
+            hits += 1
+
+    fracs[c_js] = [hits, total]
+
+
+    if do_stop:
+        break
+
+
+
+ratios = dict()
+for x,y in fracs.items():
+    ratios[x] = y[0]/y[1]
+# %%
+import math
+dir_created = set()
+for fname, ratio in ratios.items():
+    cr = math.floor(ratio * 10)/10
+    target_dir = os.path.join(target_path, str(cr))
+    if not os.path.exists(target_dir) and target_dir not in dir_created:
+        os.mkdir(target_dir)
+        dir_created.add(target_dir)
+    else:
+        dir_created.add(target_dir)
+        
+
+    
+    source_file = fname.replace('.json','.mp4')
+    target_file = os.path.join(target_dir, os.path.basename(source_file))
+
+    os.symlink(source_file, target_file)

models/20210701_202822.json

@@ -0,0 +1,23 @@
+{
+    "categories": [
+        {
+            "supercategory": "Aves",
+            "id": 206,
+            "name": "Archilochus colubris",
+            "new_id": 1
+        },
+        {
+            "supercategory": "Aves",
+            "id": 4493,
+            "name": "Icterus galbula",
+            "new_id": 2
+        },
+        {
+            "supercategory": "Aves",
+            "id": 403,
+            "name": "Poecile atricapillus",
+            "new_id": 3
+        }
+    ],
+    "model_type": "fasterrcnn_mobilenet_v3_large_fpn"
+}

score_in_directory.py

@@ -0,0 +1,35 @@
+import os
+import random
+from multiprocessing import Pool
+import sys
+sys.path.append('/home/thebears/Seafile/Designs/ML')
+from score_video import score_video
+
+rtpath = '/srv/ftp/hummingbird/2021'
+cmd = '/usr/bin/python3 /home/thebears/Seafile/Designs/ML/inaturalist_models/score_video.py {mp4name}'
+have_json = set()
+fnames = set()
+for di,_, fns in os.walk(rtpath):
+    for fn in fns:
+        if fn.endswith('.mp4'):
+            fnames.add(os.path.join(di,fn))
+        elif fn.endswith('.json'):
+            have_json.add(os.path.join(di,fn.replace('.json','.mp4')))
+
+files_to_score = list(fnames - have_json)
+random.shuffle(files_to_score)
+
+
+def try_catch_chunk(vids):
+    try:
+        score_video(vids)
+    except Exception as e:
+        print(e)
+
+lst = files_to_score
+n = 25
+chunks = [lst[i:i + n] for i in range(0, len(lst), n)]
+# %%
+if __name__ == '__main__':
+    with Pool(4) as p:
+        output = p.map(score_video,chunks)

score_video.py

@@ -0,0 +1,129 @@
+
+
+import torchvision
+from torchvision.models.detection.faster_rcnn import FastRCNNPredictor
+from collections import defaultdict as ddict
+import json
+import torch
+from torchvision import datasets, transforms as T
+import numpy as np
+import os
+import sys
+sys.path.append('/home/thebears/Seafile/Designs/ML')
+import json
+import cv2
+import random
+
+from model import Model
+import socket
+
+
+
+#vid_path = '/home/thebears/data/hummingbird_videos/Hummingbird_01_20210601055009.mp4'
+
+def score_video(vid_in_list):
+
+    
+    no_cuda =  socket.gethostname() == 'tree'
+    device='cpu'
+    model_rt_path = '/home/thebears/Seafile/Designs/ML/inaturalist_models/models/'#0210701_202822.json
+    newest_model = os.path.join(model_rt_path, max(os.listdir(model_rt_path)).replace('.pth',''))
+    with open(newest_model + '.json','r') as nmj:
+        model_json = json.load(nmj)
+
+    cats = model_json['categories']
+    cats.sort(key=lambda x: x['new_id'])
+    num_cat = len(cats) + 1
+    model_type = model_json['model_type']
+    model = Model(num_cat, model_type)
+    labels = [x['name'] for x in cats]
+    model.load_state_dict(
+        torch.load(newest_model + '.pth', map_location = torch.device(device))
+    )
+    model.eval()
+
+    if isinstance(vid_in_list, str):
+        vid_in_list = [vid_in_list]
+
+    for idx_vid, vid_in in enumerate(vid_in_list):
+        vid_path = os.path.abspath(vid_in)
+        scores_json = vid_path.rsplit('.')[0]+'.json'
+        print(os.getpid(),':',str(idx_vid),'/',str(len(vid_in_list)),vid_path)
+        if os.path.exists(scores_json):
+            print(f"JSON {scores_json} already exists")
+            exit()
+        vid_dir = os.path.dirname(vid_path)
+        os.system(f'sudo chmod 777 {vid_dir}')
+
+
+
+        cap = cv2.VideoCapture(vid_path)
+
+        from torchvision.utils import draw_bounding_boxes
+        import torch as t
+        import matplotlib.pyplot as plt
+        import matplotlib
+        cap = cv2.VideoCapture(vid_path)
+        frame_num = 0
+
+        results = list()
+        total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
+
+
+        for frame_num in range(0, total_frames, 30):
+            cap.set(cv2.CAP_PROP_POS_FRAMES, frame_num)
+            img = cap.read()[1]
+            image = img[:, :, ::-1].copy()
+            o = T.ToTensor()(image)
+            img = o[None, :, :, :]
+
+            with torch.no_grad():
+                ou = model(img)
+
+            if len(ou) > 0:
+                ofscore = ou[0]
+
+                for k in ofscore:
+                    ofscore[k] = ofscore[k].numpy().tolist() 
+
+                ofscore['names'] = [labels[x-1] for x in ofscore['labels']]
+                ofscore['frame_number'] = frame_num
+
+                results.append(ofscore)
+
+        with open(scores_json,'w') as jj:
+            json.dump(results, jj, indent=4)
+
+        
+
+if __name__ == '__main__':
+    score_video(sys.argv[1])
+
+
+# %%
+# vid_path = '/srv/ftp/hummingbird/2021/06/27/Hummingbird_01_20210627101803.mp4'
+# import time
+# import cv2
+# video = cv2.VideoCapture(vid_path)
+
+# total_frames = int(video.get(cv2.CAP_PROP_FRAME_COUNT))
+# # %%
+# st = time.time()
+
+# while True:
+#     ret, read = video.read()
+#     if not ret:
+#         break
+
+# et = time.time()
+
+# print(et-st)
+
+# st = time.time()
+# frs = list()
+# for i in range(0,total_frames, 150):
+#     video.set(cv2.CAP_PROP_POS_FRAMES, i)
+#     ret, frame = video.read()
+#     frs.append(frame)
+# et = time.time()
+# print(et-st)

test.py

@@ -0,0 +1,123 @@
+# %%
+import torchvision
+from torchvision.models.detection.faster_rcnn import FastRCNNPredictor
+from collections import defaultdict as ddict
+import json
+import torch
+from torchvision import datasets, transforms as T
+import numpy as np
+import os
+import sys
+import json
+import cv2
+import random
+from model import Model
+from torchvision.utils import draw_bounding_boxes
+import torch as t
+import matplotlib.pyplot as plt
+
+
+
+
+
+device = 'cpu'
+
+model_rt_path = '/home/thebears/Seafile/Designs/ML/inaturalist_models/models/'#0210701_202822.json
+newest_model = os.path.join(model_rt_path, max(os.listdir(model_rt_path)).replace('.pth',''))
+with open(newest_model + '.json','r') as nmj:
+    model_json = json.load(nmj)
+
+cats = model_json['categories']
+cats.sort(key=lambda x: x['new_id'])
+num_cat = len(cats) + 1
+model_type = model_json['model_type']
+model = Model(num_cat, model_type)
+labels = [x['name'] for x in cats]
+model.load_state_dict(
+    torch.load(newest_model + '.pth', map_location=torch.device(device))
+)
+model.eval()
+model.to(device)
+
+
+
+
+
+
+#rtdir = '/home/thebears/data/hummingbird_imagenet/hummingbird'
+
+#ff = [x for x in os.listdir(rtdir) if x.endswith('.jpg')]
+
+# img = os.path.join(rtdir, random.choice(ff))
+# image = cv2.imread(img)[:, :, ::-1].copy()
+# o = T.ToTensor()(image).to(device)
+# img = o[None, :, :, :]
+
+# ou = model(img)
+
+#oimage = t.tensor(image, dtype=t.uint8).permute([2, 0, 1])
+
+#matplotlib.use('Qt5Agg')
+
+
+
+
+#vid_path = '/home/thebears/data/hummingbird_videos/Hummingbird_01_20210601055009.mp4'
+#vid_path = '/home/thebears/data/hummingbird_videos/Hummingbird_01_20210617113038.mp4'
+
+print('model loaded')
+# %%
+vid_path = '/home/thebears/data/hummingbird_videos/Hummingbird_01_20210617113038.mp4'
+cap = cv2.VideoCapture(vid_path)
+imgs = list()
+
+#movie = cv2.VideoWriter('/home/thebears/Seafile/Designs/ML/inaturalist_models/output.avi',cv2.VideoWriter_fourcc('M','J','P','G'), 10, (2560,1920))
+
+frame_num = 0
+# %%
+while cap.isOpened():
+    ret, frame = cap.read()
+    if not ret:
+        break
+
+#    if frame_num % 10 == 1:
+    img = cap.read()[1]
+    image = img[:, :, ::-1].copy()
+    o = T.ToTensor()(image).to(device)
+    img = o[None, :, :, :]
+
+    ou = model(img)
+
+    idx = ou[0]['labels']
+    label_names = [labels[x-1] for x in idx]
+
+
+    scores = ou[0]['scores']
+    oimage = t.tensor(255*img.squeeze(), dtype=t.uint8)
+    boxes = ou[0]['boxes']
+
+    if boxes.shape[0] > 1:
+        boxes = boxes[[1],:]
+        label_names = [label_names[0]]
+
+    if boxes.shape[0] > 0:
+        label_names[0] += ' {0:0.2f}'.format(scores[0])
+            
+        
+    ox = draw_bounding_boxes(oimage, boxes,  width=5, labels = label_names,
+                             font='Victor Mono SemiBold Nerd Font Complete Mono Windows Compatible',font_size=50, fill = False, colors = (255, 255, 100, 100))
+    fname = '/home/thebears/Seafile/Designs/ML/inaturalist_models/frames/frame_{0:06g}.jpg'.format(frame_num)
+    from PIL import Image
+    im = Image.fromarray(np.uint8(ox.permute([1,2,0]).numpy()))
+    im.save(fname)
+
+#    plt.imshow(ox.permute([1, 2, 0]))
+    frame_num += 1
+    print(frame_num)
+
+
+
+# %%
+
+
+